A Morphological-Based License Plate Location

Abstract

License plate location is an important phase in vehicle license plate recognition for intelligent transport systems. This paper presents a robust and real time method of license plate location. The proposed algorithm consists of some stages. In the first stage, we extract vertical edges of the input image using Sobel mask. In the next stage, histogram analysis is used for finding the candidate regions of license plate. Candidate regions are also verified by defined compact factor. In the last stage, we locate the license plate exactly with some morphological operators. Experiments have been conducted for this algorithm. 400 images taken from various scenes were employed, including diverse angles, different lightening conditions. The algorithm can quickly and correctly detect the region of license plate. The license plate detecting rate of success is 83.50%.